Web Freedom Toolkit

Universal Server-Side Web Freedom Toolkit. Harmonizes Scrapling (Self-Healing Fetch), curl_cffi (TLS Impersonation), and DrissionPage (D-Mode) for undetectab...

This package appears to implement advanced, high‑privilege tooling for bypassing anti‑bot protections (CDP takeovers, driver injection, local tunnel forwarding). That capability can be legitimate for research but is risky. Before installing or running: 1) Verify whether the registry actually enforces disable-model-invocation (SKILL.md/_meta.json claim disable-model-invocation:true but registry flags here show false). If autonomous invocation is permitted, assume the agent could run high‑privilege code. 2) Inspect and understand the gating mechanisms: the code expects a lock file (~/.openclaw/tmp/sota_active.lock), a Unix socket (/tmp/.sota_auth.sock), and an env var SOTA_NUCLEAR_CONFIRMED for ‘nuclear’ actions — none are declared in manifest. 3) Only run in an isolated VM/container with no sensitive host services reachable; the relay/tunnel and CDP takeover code can be used to pivot to other local services. 4) If you do not need driver injection or socket tunneling, prefer a minimal tool that only uses scrapling or curl_cffi. 5) If you trust the author and need this functionality, require a registry-level policy that enforces disable-model-invocation:true and document the explicit human approval workflow for the nuclear paths. If any of these checks fail or you are uncomfortable, do not install or run this skill.

Type: OpenClaw Skill Name: web-freedom-toolkit Version: 8.0.0 The bundle is a highly sophisticated web scraping toolkit designed to bypass advanced anti-bot protections using DrissionPage, curl_cffi, and Scrapling. It contains several high-risk components, including a local TCP relay (python_relay.py) that forwards traffic between ports 9223 and 9222, and scripts for raw Chrome DevTools Protocol (CDP) takeover (force_takeover.py, nuclear_option.py). While the toolkit includes custom security 'gates' such as Unix Domain Socket handshakes (sota_core.py) and human-interactive challenges (secure_wrapper.py), the ability to perform low-level browser injection and maintain local listeners constitutes a high-risk capability. There is no explicit evidence of data exfiltration or malicious intent, but the 'offensive' framing and 'nuclear' options for bypassing security flags justify a suspicious classification.